Experimental study on oxygen concentrator with wide product flow rate range: individual parametric effect and process improvement strategy

https://doi.org/10.1016/j.seppur.2021.118918Get rights and content

Highlights

  • Flexible improvement on PSA O2 production at wide product flow rate range is studied.

  • Individual effects of process parameters on O2 production performance are obtained.

  • Adjusted/unadjusted PSA experiments using appropriate valve settings are conducted.

  • Process improvement strategies for different product flow rate ranges are proposed.

  • Higher O2 purity, recovery and productivity with lower energy consumption are gained.

Abstract

Improvement on oxygen (O2) concentrator using pressure swing adsorption (PSA) technology according to variable product demands is of great significance, which, for instance, provides the key step for success in practical O2 therapy for COVID-19 patients who need to be delivered with a wide product flow rate range (1–15 standard liter per minute (SLPM)) of medical O2 (purity > 82%). This work studied the individual effects of major PSA process parameters on O2 production performance at the product flow rate of 3.46–19.88 SLPM (0.64–3.68 SLPM per kilogram of adsorbent (SLPM/kg)), based on a self-designed two-bed PSA unit with a modified Skarstom cycle using Li-LSX zeolite adsorbents. The improvement strategies were accordingly proposed based upon influential mechanisms of each parameter: 1) at lower product flow rates (≤2.00 SLPM/kg), increasing the purge flow rate and decreasing the adsorption pressure to suppress excess O2 adsorption, and decreasing the feed flow rate to ensure low energy consumption; 2) at higher product flow rates (≥2.00 SLPM/kg), decreasing the purge flow rate and increasing the adsorption pressure to eliminate N2 breakthrough and O2-rich product waste, and appropriately increasing the feed flow rate to enhance cost-effectiveness. An improved set of parameters rendered O2 purity (95.67–74.86%), recovery (11.28–49.05%), productivity (0.47–2.04 mmol/kg/s) increased by up to 3.52–20.08%, 0.36–20.47%, 6.82–19.61%, and energy consumption (4.07–0.95 kWh/kgO2) decreased by up to 10.56–18.10%, in comparison to two conventional sets, respectively. The results are beneficial for developing intellectualized and flexibly-controlled O2 concentrators for practical applications.

Introduction

In the past few decades, the oxygen (O2) concentrator using pressure swing adsorption (PSA) technology plays a significant role in indoor and medical applications, due to its advantages of cost-effectiveness, high O2 purity, safety, and negligible pollution [1], [2]. During practical applications of O2 concentrators, the product flow rate varies with actual requirements. For example, the supply of medical O2 recommended for severe and critical COVID-19 patients involves low doses ranging from 1 to 2 SLPM (standard liters per minute) for children and starting at 5 SLPM for adults with nasal cannula, moderate flow rates of 6–10 SLPM for venturi mask and higher flow rates of 10–15 SLPM for the mask with a reservoir bag, according to the interim guidance issued by the World Health Organization (WHO) [3]. However, commercial O2 concentrators that generally deliver flow rates of 0.5–10 SLPM [3] barely meet the need of COVID-19 treatment. It is therefore necessary to develop single-use O2 concentrators producing medical grade (≥82% pure O2 [3]) at a wide product flow range (1–15 SLPM for COVID-19 treatments). To satisfy the variable product demand, conventional PSA O2 production processes remain challenging in terms of the trade-off between product quality (O2 purity, recovery, and productivity) and energy consumption throughout the whole product range. In view of this, a flexible improvement strategy at varying product flow rate is of great importance.

The variations of O2 purity and recovery with product flow rate has been extensively studied in literatures [4], [5], [6], [7], [8], [9]. Farooq et al. found that as the product flow rate increased from 0.06 to 0.24 SLPM, the O2 recovery increased and the O2 purity maintained using 5A zeolites, speculating that the O2 purity would decrease if the product flow rate further increases [4]. Santos et al. verified Farooq's speculation showing that as the product flow rate increased from 0.96 to 6 SLPM, the O2 recovery was increased by 23% with O2 purity kept constant (~95%) but then decreased to 60% using 13X zeolites [5]. Mendes et al. [6] and Beeyani et al. [7] also found gradual decreases in O2 purities with product flow rates from 0.1 to 1.1 SLPM (0.16 to 1.78 SLPM per kilogram of adsorbent (SLPM/kg)) and from 0.06 to 0.24 SLPM, respectively, using 5A zeolites. Mofarahi et al. [8] and Bhat et al. [9] showed first increase and subsequent decrease in O2 purity with the product flow rate from 0.3 to 1.5 SLPM (0.16 to 0.79 SLPM/kg) for an experimental device using 5A zeolites and from 0 to 100 SLPM for an on-board O2 concentrator using Li-LSX zeolites, respectively. They attributed the decrease of O2 purity with product flow rate to the reduction of purge flow rate or purge gas purity, causing N2 breakthrough in the adsorption stage [8], [9], [10]. Some other works also gave explanations that N2 adsorption equilibrium may not be achieved and leads to a stretching of the mass transfer zone [11], [12], [13], [14], [15]. Nevertheless, few discussions were reported on the rise of O2 purity with product flow rate in the lower range. .

The PSA O2 production performance can be affected by the product flow rate as well as other process parameters which are in fact cross-correlated. For instance, the product flow rate affects the purge flow rate, the adsorption pressure, and the feed flow rate, which associate to bed regeneration [10], adsorbent utilization and N2 adsorption capacity [6], [7], respectively. Influential mechanisms of different process parameters on O2 purity and recovery with varying product flow rates remain unclear because previous PSA O2 production experiments were mostly conducted with fixed product flow rates or uncontrolled process parameters at fixed valve settings [4], [5], [6], [7], [8], [9], [16], [17], [18]. Studies on the effect of a parameter on O2 production were always coupled with changes in other correlated parameters. Therefore, the individual parametric effect on PSA O2 production merits deeper investigation that may require specifically designed experimental methodology.

In this work, the precise adjustment and control of each key parameter (product flow rate, purge flow rate, feed flow rate, and adsorption pressure) were enabled by appropriate needle valve setting within the PSA apparatus. Based on unadjusted and adjusted experiments using a modified Skarstrom cycle with Li-LSX zeolite adsorbents, the individual parametric effect on O2 production performance at wide product flow rate range (3.46–19.88 SLPM) were studied. The improvement strategies at lower and higher product flow rates were proposed, and accordingly an improved set of process parameters were obtained with greater O2 purity, recovery, productivity at lower energy consumption.

Section snippets

Experimental setup and PSA O2 production process

The experimental setup of the two-bed (B1 and B2) PSA O2 production system is shown in Fig. 1 along with the apparatus physical dimensions listed in Table 1. A layered adsorbent bed was used, consisting of the pre-layer activated alumina for water vapor removal [19] and the main-layer Li-LSX zeolites for N2/O2 separation [9], [20]. The activated alumina (model JLAA-1) and Li-LSX zeolites (model JLOX-101A) were supplied by Luoyang Jalon Micro-nano New Materials Co., Ltd. The adsorbent physical

Effects of Qp on O2 purity and recovery

Fig. 3 shows the results of variations of the O2 purity and recovery with Qp in unadjusted experiments A and B. As Qp increased, the O2 purity dramatically declined from 95.11 to 59.83% in experiment A, while it firstly increased, peaked over 95% at 1.92 SLPM/kg, and then decreased in experiment B. The two types of O2 purity variations are consistent with earlier work from Mendes et al. [6] and Beeyani et al. [7], and from Mofarahi et al. [8] and Bhat et al. [9], respectively. The O2 recovery

Conclusion

In this work, the individual effects of PSA process parameters on O2 production performance in a wide product flow rate range were systematically studied with unadjusted and adjusted experiments based on a specially designed two-bed PSA methodology. It was found that the negative effects of EOA at lower product flow rates and of N2 breakthrough at higher product flow rates on O2 production performance may be weakened or eliminated by adjusting the purge flow rate, the feed flow rate and the

CRediT authorship contribution statement

Quanli Zhang: Conceptualization, Investigation, Methodology, Writing - original draft. Yingshu Liu: Funding acquisition, Supervision. Ziyi Li: Conceptualization, Funding acquisition, Data curation, Writing - review & editing. Penny Xiao: Writing - review & editing. Wenhai Liu: Investigation. Xiong Yang: Formal analysis. Yaoguo Fu: Investigation. Chunyu Zhao: Investigation. Ralph T. Yang: Validation. Paul A. Webley: Validation.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This paper obtained the support from the National Key R&D Program of China (No. 2020YFC1512302, 2017YFC0806304), the National Natural Science Foundation of China (No. 21808012, 21676025), and the Fundamental Research Funds for the Central Universities (FRF-IDRY-19–025, FRF-TP-20-011A2)

References (36)

  • W.H.O. Who

    Technical specifications for Pressure Swing Adsorption

    (2020)
  • W.H.O. WHO, Oxygen sources and distribution for COVID-19 treatment centres,Interim guidance....
  • S.D. Farooq et al.

    Numerical simulation of a pressure swing adsorption oxygen unit

    Chem. Eng. Sci.

    (1998)
  • J.C. Santos et al.

    Simulation and Optimization of Small Oxygen Pressure Swing Adsorption Units

    Ind. Eng. Chem. Res.

    (2004)
  • A.K. Beeyani et al.

    Parametric studies and simulation of PSA process for oxygen production from air

    Pol J Chem Technol.

    (2010)
  • M. Mofarahi et al.

    Oxygen Separation from Air by Four-Bed Pressure Swing Adsorption

    Ind. Eng. Chem. Res.

    (2009)
  • A.A. Bhat et al.

    On-Board Oxygen Generation Using High Performance Molecular Sieve

    Defence Life Science Journal.

    (2017)
  • A.S.T. Chiang et al.

    Experimental study on a four-bed PSA air separation process

    AIChE J.

    (1994)
  • Cited by (13)

    • Quantitative analysis of droplet deposition produced by an electrostatic sprayer on a classroom table by using fluorescent tracer

      2021, Building and Environment
      Citation Excerpt :

      For those reasons, currently, surface decontamination of shared surfaces is officially recommended by several authorized organizations, such as WHO [6], ECDC [7], OSHA [8], and CDC [9]. However, most recent studies are solely focused on droplets or aerosols transmission to mitigate the virus transmission [10–17]. As main surface decontamination methods like surface wiping, trigger sprayers, pressure sprayers, and electrostatic sprayers are used with disinfectant solutions for various places from households to public places such as schools, public transportations, and hospitals [18,19].

    View all citing articles on Scopus
    View full text